Center for Computational and Theoretical Biology
Refine
Has Fulltext
- yes (2) (remove)
Is part of the Bibliography
- yes (2)
Year of publication
- 2022 (2) (remove)
Document Type
- Journal article (2)
Language
- English (2)
Keywords
- AI (1)
- FLIMbee (1)
- HDBSCAN (1)
- MiMIC (1)
- SMLM (1)
- STORM (1)
- Unc-13 (1)
- active zone (1)
- compressed sensing (1)
- dSTORM (1)
Institute
- Center for Computational and Theoretical Biology (2)
- Theodor-Boveri-Institut für Biowissenschaften (2)
- Klinik und Poliklinik für Psychiatrie, Psychosomatik und Psychotherapie (1)
- Klinik und Poliklinik für Unfall-, Hand-, Plastische und Wiederherstellungschirurgie (Chirurgische Klinik II) (1)
- Physiologisches Institut (1)
EU-Project number / Contract (GA) number
- 835102) (1)
Introduction
Neurotransmitter release at presynaptic active zones (AZs) requires concerted protein interactions within a dense 3D nano-hemisphere. Among the complex protein meshwork the (M)unc-13 family member Unc-13 of Drosophila melanogaster is essential for docking of synaptic vesicles and transmitter release.
Methods
We employ minos-mediated integration cassette (MiMIC)-based gene editing using GFSTF (EGFP-FlAsH-StrepII-TEV-3xFlag) to endogenously tag all annotated Drosophila Unc-13 isoforms enabling visualization of endogenous Unc-13 expression within the central and peripheral nervous system.
Results and discussion
Electrophysiological characterization using two-electrode voltage clamp (TEVC) reveals that evoked and spontaneous synaptic transmission remain unaffected in unc-13\(^{GFSTF}\) 3rd instar larvae and acute presynaptic homeostatic potentiation (PHP) can be induced at control levels. Furthermore, multi-color structured-illumination shows precise co-localization of Unc-13\(^{GFSTF}\), Bruchpilot, and GluRIIA-receptor subunits within the synaptic mesoscale. Localization microscopy in combination with HDBSCAN algorithms detect Unc-13\(^{GFSTF}\) subclusters that move toward the AZ center during PHP with unaltered Unc-13\(^{GFSTF}\) protein levels.
Background
Localization-based super-resolution microscopy resolves macromolecular structures down to a few nanometers by computationally reconstructing fluorescent emitter coordinates from diffraction-limited spots. The most commonly used algorithms are based on fitting parametric models of the point spread function (PSF) to a measured photon distribution. These algorithms make assumptions about the symmetry of the PSF and thus, do not work well with irregular, non-linear PSFs that occur for example in confocal lifetime imaging, where a laser is scanned across the sample. An alternative method for reconstructing sparse emitter sets from noisy, diffraction-limited images is compressed sensing, but due to its high computational cost it has not yet been widely adopted. Deep neural network fitters have recently emerged as a new competitive method for localization microscopy. They can learn to fit arbitrary PSFs, but require extensive simulated training data and do not generalize well. A method to efficiently fit the irregular PSFs from confocal lifetime localization microscopy combining the advantages of deep learning and compressed sensing would greatly improve the acquisition speed and throughput of this method.
Results
Here we introduce ReCSAI, a compressed sensing neural network to reconstruct localizations for confocal dSTORM, together with a simulation tool to generate training data. We implemented and compared different artificial network architectures, aiming to combine the advantages of compressed sensing and deep learning. We found that a U-Net with a recursive structure inspired by iterative compressed sensing showed the best results on realistic simulated datasets with noise, as well as on real experimentally measured confocal lifetime scanning data. Adding a trainable wavelet denoising layer as prior step further improved the reconstruction quality.
Conclusions
Our deep learning approach can reach a similar reconstruction accuracy for confocal dSTORM as frame binning with traditional fitting without requiring the acquisition of multiple frames. In addition, our work offers generic insights on the reconstruction of sparse measurements from noisy experimental data by combining compressed sensing and deep learning. We provide the trained networks, the code for network training and inference as well as the simulation tool as python code and Jupyter notebooks for easy reproducibility.